Skip to end of banner
Go to start of banner

Modules

Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

Version 1 Next »

With Gnoppix 24.4 we have 6 modules

  • a 3.6GB module called gnoppix-uncensored-fast 7 Billion parameter 4bit

  • a 6.9GB module called gnoppix-uncensored-huge 13 Billion parameter 4bit

  • a 1.5GB gnoppix-code-v1 7 Billion parameter 4bit

  • a 3.6GB gnoppix-code-PRO 13 Billion parameter 4bit

  • a 3.8GB opensource mistral modul 7 Billion parameter 4bit

  • a 3.6GB opensource wizard-vicuna-uncensored modul 7 Billion parameter 4bit

In general, module gnoppix-uncensored-fast, answer super fast less than 1sec. I’ve added also

coding modules which help me generating / debugging code (smile) After trying and optimizing a lot I think

the module can answer you EVERY question with a professional output.

Q: Why there are multiple modules, isn't just 1 enough?

A: I’m using this also for image creation, with Gnoppix 24.5 you can create and modify images with AI. LibreOffice is also prove reading or generating your documents.

Important Note: 

An uncensored model has no guardrails. 

You are responsible for anything you do with the model, just as you are responsible for anything you do with any dangerous object such as a knife, gun, lighter, or car.

Publishing anything this model generates is the same as publishing it yourself.

You are responsible for the content you publish, and you cannot blame the model any more than you can blame the knife, gun, lighter, or car for what you do with it. If you feel uncomfortable with it, do NOT use it.

Shout out to the open source AI/ML community, and everyone who helped to train the models.

Wizard Vicuna Uncensored is a 7B parameter model based on Llama 2 uncensored by Eric Hartford. The models were trained against LLaMA-7B with a subset of the dataset, responses that contained alignment / moralizing were removed.

family: llama parameters:7B quantization: 4-bit

  • No labels